Antonio Cavallaro Salvatore Grasso Michele Maugeri CNR Ibam, Catania University of Catania University of Catania

Transcription

1 DYNAMIC GEOTECHNICAL CHARACTERIZATION OF SAN GIULIANO DI PUGLIA SEISMIC AREA Antonio Cavallaro Salvatore Grasso Michele Maugeri CNR Ibam, Catania University of Catania University of Catania Viale A. Doria, Viale A. Doria, Viale A. Doria, 9 Catania 9 Catania 9 Catania ABSTRACT The city of San Giuliano di Puglia (CB), located in the Molise region in Southern Italy, is prone to high seismic risk. For site characterisation of soil deep site investigations have been undertaken. Borings, static and dynamic in situ tests have been performed. Among them Cone Penetration Tests (CPT), Cross-Hole (C-H) Down-Hole (D-H) and Seismic Dilatometer Marchetti Tests (SDMT) have been carried out, with the aim to evaluate the soil profile of shear waves velocity (V s ). Moreover the following laboratory tests were carried out on undisturbed samples: Oedometer tests, undrained Triaxial tests and Resonant Column tests. The available data enabled one to compare the shear waves velocity profile obtained by empirical correlations, Down Hole tests and Seismic Dilatometer Marchetti Tests. The influence of strain level on G-γ and D-γ curves was evaluated by means of laboratory tests. Two expression to allows the complete shear modulus degradation with strain level and the inverse variation of damping ratio with normalized shear modulus respectively were proposed. Finally after evaluating the synthetic accelerograms at the bedrock, the ground response analysis at the surface, in terms of time history and response spectra, has been obtained by two non-linear models GEODIN and EERA. INTRODUCTION On October 3,, a M L =. earthquake struck Molise region in Southern Italy. The strongly non-uniform damage distribution observed in the town of San Giuliano di Puglia suggested that site amplification significantly affected the seismic response of the area. The damages were surprisingly concentrated in the new part of the town, lying on fine-grained soils, where a primary school building collapsed causing the death of 7 children. While the old part, lies on outcropping rock, was less damaged. In order to develop a seismic microzonation of the area (Baranello et al. 3), the Department of Civil Protection of the Italian Government committed a comprehensive investigation programme on the subsoil properties. The field and laboratory experimental data allowed the definition of the geotechnical model of subsoil (Silvestri et al., ), while a numerical site response analyses were carried out by Puglia et al. (7). This paper tries to summarize this information in a comprehensive way in order to provide a case record of site characterization for seismic response analysis. GEOLOGY OF AREA Melidoro () and Guerricchio () have proposed an interpretation of the town geological setting of San Giuliano di Puglia area (Figure ): - the Faeto flysch (F), that is a sedimentary succession of mainly calcareous soils, either coarse or finegrained; in particular these can be calcirudites, limestones, calcareous marls, white marls and green clays, differently fractured and fissured; - a deep layer of Toppo Capuana marly clays (MC), whose maximum thickness has not been assessed yet; at the top, these clays are weathered down to few meters; they are overtopped by a shallow cover of disturbed soil and landslide debris; - a chaotic complex (C) formed by Varicolored Scaly clays, limestones, calcareous marls, calcarenites and fragments of Faeto flysch. As shown in Figure, the MC formation is in lateral contact with formation F, that emerges in the Southern part of ridge, where the flysch appears to be less fractured and constitutes the foundation soil of the historical part of the town, the Faeto flysch is heavily tectonized and broken up. The Toppo Capuana marly clay formation at San Giuliano di Puglia consists of three principal units: - a debris cover, of less than five meters thickness, including black organic carbonaceous elements, Paper No. 7.a

2 lumps and lenses of white powdery calcite and small calcareous litho-clasts; - a layer, of two to ten meters thickness, of weathered tawny clays, characterized by medium to intense fissuring, resulting from the weathering and disturbance of the uppermost part of Toppo Capuana marly clays; - a deep layer of Toppo Capuana marly clays, called grey clays hereafter. The thickness of this layer seems to be around three hundred meters. The grey Toppo Capuana marly clays are less intensely fissured than the weathered tawny clays; in some cases the fissure surfaces are either ochraceous or covered by a black oxidation patina. The polyhedral clay elements, of cm maximum size, are sharply edged and well embedded within the fissure network. Crystals of selenitic gypsum are detectable in rare lenses or in thin layers of fine sand. BASIC GEOTECHNICAL SOIL PROPERTIES Boreholes driven to a depth of m were performed and undisturbed samples were retrieved for laboratory tests. The testing programme consisted of standard classification tests, Oedometer tests, undrained triaxial tests and resonant column tests. The general characteristics and index properties of the Toppo Capuana marly clay formation at San Giuliano di Puglia are shown, as a function of depth, in figure. The overall values of the main physical properties are summarized in Table, which reports the average values. The values of the natural moisture content w n prevalently range between 7 and %. Characteristics values for the Atterberg s limits are: w l = 3-3 % and w p = 3 - %, with a plasticity index of PI = 3 - % (Silvestri et al., ; Vitone, ). Table. Physical average properties of the three units. Debris Cover Tawny Clay Grey Clay Thickness m m 3 m Sand Fraction [%]. 3.. Silt Fraction [%]... Clay Fraction [%].7.. γ [KN/m 3 ] e [-].7..9 G s [-] w n [%] w p [%] w l [%] c' [kpa] 3 3 φ' [ ] Fig.. Geological map and N-S section of San Giuliano di Puglia (after Silvestri et al., ). where: c' (Cohesion) and φ' (Angle of shear resistance) were calculated from C-U Triaxial Tests. The data shown in figure clearly indicate a good degree of homogeneity of the deposit. This indication is also confirmed by comparing the penetration resistance q c from mechanical cone penetration tests (CPT) performed at different locations over the investigated area (Figure 3). The variation of q c with depth clearly shows the existence of layers with very different mechanical characteristics. The upper debris cover with q c of about. to. MPa. The weathered tawny clays has q c values of about 3. to.7 MPa. The grey clays deep layer has q c values of about.9 to. MPa. Two transition zones (q c. to.7 MPa) exists between these three strata at depths of about and m. Paper No. 7.a

3 CF (%) 3 3 Wp - Wn - Wl (%) IC.... γ (kn/m 3 ) 9 e depth [m] Wp Wn Wl Fig.. Index properties of Toppo Capuana marly clays. H [m] q c [MPa] Fig. 3. Static cone penetration test results. S.Giuliano di Puglia CPT SHEAR MODULUS AND DAMPING RATIO FROM LABORATORY TESTS Shear modulus G and damping ratio D of Toppo Capuana marly clay formation were obtained in the laboratory from Resonant Column tests (RCT). The laboratory test conditions and the obtained small strain shear modulus G o are listed in Table. The undisturbed specimens were isotropically reconsolidated to the best estimate of the in situ mean effective stress. The size of solid cylindrical specimens are Radius = mm and Height = mm. S S S Table. Test Condition for Toppo Capuana marly clay formation specimens. σ vc Borehole H e PI RCT G o U max No. [m] [kpa] [MPa] [kpa] S3C U 3 SC U 3 SC U 33 9 SC U 73 SC U where: U = Undrained. G/G o..... S3 - C S - C S - C S - C3 S - C Yokota et al. (9)..... γ [%] Fig.. G/G o -γ curves from RCT tests. Figure shows the results of RCTs normalized by dividing the shear modulus G(γ) for the initial value G o at very low strain. Paper No. 7.a 3

4 The experimental results of specimens from Toppo Capuana marly clay formation were used to determine the empirical parameters of the eq. proposed by Yokota et al. (9) to describe the shear modulus decay with shear strain level: G(γ) = () β G o + αγ(%) in which: G(γ) = strain dependent shear modulus; γ = shear strain; α, β = soil constants. The expression () allows the complete shear modulus degradation to be considered with strain level. The values of α = and β =. were obtained for Toppo Capuana marly clay formation. D [%] S3 - C S - C S - C S - C3 S - C Yokota et al. (9).... G/Go Fig.. D-G/G o curves from RCT tests. As suggested by Yokota et al. (9), the inverse variation of damping ratio with respect to the normalized shear modulus has an exponential form as that reported in Figure for the central area of Catania: ( γ) G D( γ)(%) = η exp λ () G o in which: D(γ) = strain dependent damping ratio; γ = shear strain; η, λ = soil constants. The values of η = and λ =. were obtained for Toppo Capuana marly clay formation. The equation () assume maximum value D max = % for G(γ)/G o = and minimum value D min =.9 % for G(γ)/G o =. Therefore, eq. () can be re-written in the following normalized form: D(γ) D(γ) G = exp λ max G o ( γ) EVALUATION OF G O FROM PENETRATION TESTS It was also attempted to evaluate the small strain shear modulus by means of the following empirical correlations based on penetration tests results or laboratory results available in literature. a) Hryciw (99): G 3 γ / γ D w. '. o = Ko ( σ '. v pa ( σ.7 ) v/pa ) γ D/ γ w where: G o, σ v and p a are expressed in the same unit; p a = bar is a reference pressure; γ D and K o are respectively the unit weight and the coefficient of earth pressure at rest, as inferred from DMT results according to Marchetti (9). b) Mayne and Rix (993): G o q = 3. e. 9 c where: G o and q c are both expressed in [kpa] and e is the void ratio. Eq. () is applicable to clay deposits only. c) Jamiolkowski et. al. (99): G o = e σ '. m p. a. 3 where: σ m = (σ v + σ h ) / 3; p a = bar is a reference pressure; G o, σ m and p a are expressed in the same unit. The values for parameters which appear in eq. () are equal to the average values that result from laboratory tests performed on quaternary Italian clays and reconstituted sands. A similar equation was proposed by Shibuya and Tanaka (99) for Holocene clay deposits. Eqs. () and () incorporate a term which expresses the void ratio; the coefficient of earth pressure at rest only appear in eq. (). However only eq. () tries to obtain all the input data from the DMT results. The G o values obtained with the methods above indicated are plotted against depth in Figure. The method by Jamiolkowski et al. (99) was applied considering a given profile of void ratio and K o. The coefficient of earth pressure at rest was inferred from DMT. The methods by Hryciw (99) and Jamiolkowski et. al. (99) show very different G o values of the soil respect to the method by Mayne and Rix (993). On the whole, equation () seems to provide the most accurate trend of G o with depth, as can be seen in Figure. The DMT material index was not (3) () () () Paper No. 7.a

5 capable of detecting the presence of different strata at depths of about and m. H [m] 3 7 Hryciw (99) Jamiolkowski (99) Mayne and Rix (993) S Mayne and Rix (993) S Mayne and Rix (993) S G o [MPa] Fig.. G o from different empirical correlations. SHEAR MODULUS FROM IN SITU TESTS Values of G o have been obtained from Down-Hole (D-H) and Seismic Dilatometer Marchetti Test (SDMT). The SDMT provides a simple means for determining the initial elastic stiffness at very small strains and in situ shear strength parameters at high strains in natural soil deposits. Location of SDMT tests at San Giuliano di Puglia site is shown in Figure 7. Figure shows the values of G o obtained in situ from D-H tests and Seismic Dilatometer Marchetti Test (SDMT). While always in the Figures and the results by D-H tests are comparable with those obtained by Hryciw (99) and Jamiolkowski et. al. (99) correlations. H [m] 3 3 G o [MPa] 3 7 Fig.. G o from in situ tests. S. Giuliano di Puglia DH DH DH 7 DH DH 9 DH DH SDMT GROUND RESPONSE ANALYSIS AT THE SOIL SURFACE USING DIFFERENT V S PROFILES The site response was made by -D non-linear computer code GEODIN. The code implements a one-dimensional simplified, hysteretic model for the non-linear soil response (Frenna and Maugeri, 99). The S-wave propagation obtained by D-H and SDMT occurs on a -D column having shear behavior. The column is subdivided in several, horizontal, homogeneous and isotropic layers characterized by a non-liner spring stiffness G(γ), a dashpot damping D(γ) and a soil mass density ρ. Moreover, to take into account the soil non-linearity, laws of shear modulus and damping ratio against strain have been inserted in the code. The -D columns have a height of 3 m and are excited at the base by accelerograms obtained from the recording of the aftershocks of the earthquake of October 3, (M L =.; Mw =.7) (Lat..7; Long..9) (see Figure 9). The analysis provides the time-history response in terms of displacements, velocity and acceleration at the surface. Using this time history, response spectra concerning the investigated site have been deduced. Fig. 7. Location of SDMT tests at San Giuliano di Puglia. The G o values are plotted in Figure against depth. It is possible to see that quite a good agreement exists between the D-H tests results. Higher values of G o was obtained by SDMT. Considering the Figures and 7 it is possible to see a good agreement between the G o results obtained by Mayne and Rix (993) and by in situ SDMT. Fig. 9. Acceleration (g) of the recorded aftershock used for analyses of local seismic response. Paper No. 7.a

6 The soil response at the surface was also modeled using the Equivalent linear Earthquake site Response Analyses of Layered Soil Deposits computer code EERA (Bardet et al., ) for calculus of amplitude ratios and spectral acceleration. Figure shows as example the stratigraphy of borehole S3 (near SDMT 3) with indication of V s profile. It is possible to find a layer, of about three meters thickness, of weathered tawny clays, characterized by medium to intense fissuring, resulting from the weathering and disturbance of the uppermost part of Toppo Capuana marly clays; then a deep layer of Toppo Capuana marly clays, called grey clays hereafter. V s [m/s] 3 SDMT3 Figure shows the variation with depth of maximum shear strain, ratio G/G max and damping ratio at each of the calculation iteration step performed using EERA. At the first calculation step, the ratio G/G max is equal to, and the damping ratio is constant. After a few calculation steps, the distributions of ratio G/G max and damping ratio converge toward their final values.,, Maximum Shear strain (%),9,9,9, G/G max Damping Ratio (%) H [m] Fig.. Variation with depth of maximum shear strain, ratio G/G max and damping ratio. Figure 3 shows the corresponding variation with depth of maximum shear stress during calculations, and the converged maximum acceleration. Maximum Shear Stress (kpa) 3 Maximum Acceleration (g),,,3 Fig.. Stratigraphy of borehole S3 (near SDMT 3) with indication of Vs profile. In Figure is reported the time history of accelerations at the soil surface, to allow calculus respectively of amplitude ratios and response spectra using GEODIN and EERA. Acceleration (g),,,, -, -, -, -, -, Time (sec) Fig.. Time history of acceleration at the soil surface (g) of the recorded aftershock of earthquake of October 3,. Fig. 3. Variation with depth of maximum shear stress during calculations, and the converged maximum acceleration. Figure shows the time histories of shear strain, shear stress and energy dissipated per unit volume, and the stress-strain loop computer at sub-layer No.. Figure shows the computed amplitude of amplification ratio between bottom and free surface and the computed amplitude of Fourier amplitude at free surface. Paper No. 7.a

7 S train (% ) Strain Energy (kpa),, -, -, -, Stress (kpa) Time (sec),, -, - -, ,,,,,, Stress (kpa),, -, - -, Time (sec) Time (sec) - -, -, -,,, Strain (%) Fig.. Time histories of shear strain, shear stress and energy dissipated per unit volume, and the stress-strain loop computer at sublayer No.. Fourier Amplitude Amplification Ratio,,,,, 3 Frequency (Hz) Frequency (Hz) Fig.. Computed amplitude of amplification ratio between bottom and free surface and computed amplitude of Fourier amplitude at free surface. Response spectra at the soil surface have been then obtained from GEODIN and EERA using components of the recorded earthquake. Figure displays the acceleration response spectra computed at free surface for a % critical damping ratio. Spectral Acceleration (g),,7,,,,3,, EERA GEODIN % GEODIN %,,, 3 Period (s) Fig.. Acceleration response spectra computed at free surface for a % critical damping ratio. CONCLUDING REMARKS In this paper some information concerning G and D of San Giuliano di Puglia area in the Molise region for seismic response analysis have been presented. Available data enabled one to define the small strain shear modulus profile to describe the G and D variation with strain level. Experimentally determined G o profiles from in situ tests were compared to that inferred from laboratory test results. It was then possible to evaluate the seismic response of the soil excited at the base by accelerograms obtained from the recording of the aftershocks of the earthquake of October 3,. The site response was made by -D non-linear computer code GEODIN and also by the Equivalent linear Earthquake site Response Analyses of Layered Soil Deposits computer code EERA. The analysis provided the time-history response at the surface, the computed amplitude of amplification ratio between bottom and free surface and the acceleration response spectra at the free surface. REFERENCES Baranello S., Bernabini M., Dolce M., Pappone G., Rosskopf C., Sanò T., Cara P. L., De Nardis R., Di Pasquale G., Goretti A., Gorini A., Lembo P., Marcucci S., Marsan P., Martini M. G., Naso G. [3]. Rapporto finale sulla Microzonazione Sismica del centro abitato di San Giuliano di Puglia, Department of Civil Protection, Rome, Italy. Bardet J. P., Ichii K. and Lin C. H., []. EERA: a computer program for equivalent-linear earthquake site response analyses of layered soil deposits, user manual, University of Southern California, pp. Paper No. 7.a 7

8 Frenna S. M. and Maugeri M. [99]. GEODIN: a computer code for seismic soil response, Proceeding of the. 9 th Italian Conference of Computational Mechanics, Catania, Italy, - June 99, (in Italian): -. Guerricchio []. Private communication, University of Calabria. Hryciw R.D. [99]. Small strain shear modulus of soil by Dilatometer, JGED, ASCE, Vo., N. : 7-7. Jamiolkowski M., Lo Presti D.C.F. and Pallara O. [99]. Role of in-situ testing in geotechnical earthquake engineering, Proceedings of the 3 rd International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamic, State of the Art 7, St. Louis, Missouri, April -7, 99, vol. II: 3-. Marchetti S. [9]. In situ tests by Flat Dilatometer, Journal of the Geotechnical Engineering Division, ASCE, Vol., N. GT3, March, 9: Mayne P.W. and Rix G.J. [993]. G max -q c relationships for clays, Geotechnical Testing Journal, Vol., N. : -. Melidoro G. []. Private communication, Technical University of Bari. Puglia R., Lanzo G., Pagliaroli A., Sica S. and Silvestri F. [7]. Ground motion amplification in S. Giuliano di Puglia (southern Italy) during the Molise earthquake, Proceedings of the International Conference on Earthquake Geotechnical Engineering. Shibuya S. and Tanaka H. [99]. Estimate of elastic shear modulus in holocene soil deposits, Soils and Foundations, Vol. 3, N. : -. Sivestri, F., Vitone, C., d Onofrio, A., Cotecchia, F., Puglia, R. and Santucci de Magistris, F. []. The influence of meso-structure on the mechanical behaviour of a marly clay from low to high strains, Proceedings of the Geotechnical Symposium to Celebrate Prof. Tatsuoka s th Birthday, Roma, - 7 March. Vitone, C., Cotecchia, F., Santaloia, F. and Cafaro, F. []. Preliminary results of a comparative study of the compression behaviour of clays of different fissuring, Proceedings of the International Conference on Problematic Soils, Cyprus, pp Yokota, K., Imai, T. and Konno, M. [9]. Dynamic deformation characteristics of soils determined by laboratory tests, OYO Tec. Rep. 3, Paper No. 7.a